Citations

Abstract Dysexecutive Alzheimer’s disease (dAD) manifests as a progressive dysexecutive syndrome without prominent behavioral features, and previous studies suggest clinico-radiological heterogeneity within this syndrome. We uncovered this heterogeneity using unsupervised machine learning in 52 dAD patients with multimodal imaging and cognitive data. A spectral decomposition of covariance between FDG-PET images yielded six latent factors (“eigenbrains”) accounting for 48% of variance in patterns of hypometabolism. These eigenbrains differentially related to age at onset, clinical severity, and cognitive performance. A hierarchical clustering on the eigenvalues of these eigenbrains yielded four dAD subtypes, i.e. “left-dominant,” “right-dominant,” “bi-parietal-dominant,” and “heteromodal-diffuse.” Patterns of FDG-PET hypometabolism overlapped with those of tau-PET distribution and MRI neurodegeneration for each subtype, whereas patterns of amyloid deposition were similar across subtypes. Subtypes differed in age at onset and clinical severity where the heteromodal-diffuse exhibited a worse clinical picture, and the bi-parietal had a milder clinical presentation. We propose a conceptual framework of executive components based on the clinico-radiological associations observed in dAD. We demonstrate that patients with dAD, despite sharing core clinical features, are diagnosed with variability in their clinical and neuroimaging profiles. Our findings support the use of data-driven approaches to delineate brain–behavior relationships relevant to clinical practice and disease physiology.

Purpose To analyze the current literature regarding risk factors associated with medial ulnar collateral ligament (MUCL) injury in baseball players and to serve as a robust source for identifying modifiable risk factors that once optimized, have the potential to reduce injury risk. Methods Comprehensive search of the available literature was performed according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis guidelines. Studies were included if they evaluated risk factors for MUCL injuries in the elbow of baseball players. Risk of bias assessment was performed via Methodological Index for Non‐randomized Studies (MINORS) scoring system. The Oxford Centre for Evidence‐Based Medicine was used to determine level of evidence. Variables of interest; player age, position, shoulder motion, humeral retrotorsion, joint laxity, strength, balance, geography, velocity, pitch count, pitch types, throwing volumes, and throwing mechanics were recorded. Results Twenty‐one studies were included in this systematic review. MINORS scores ranged from 75 to 87%, and variables demonstrated significant heterogeneity. Performance‐based risk factors for MUCL injury included: increased pitch count (both annual and per game), higher percentage of fastballs thrown, smaller pitch repertoire, and/or a loss of pitching velocity. Biomechanical studies demonstrated the relationship between decreased shoulder range of motion (total ROM, ER, IR, and abduction), increased humeral retrotorsion, increased elbow valgus opening in the throwing arm, lower Y‐Balance score, and increased lateral release position to increased MUCL injury. Conclusions Risk factors for MUCL injury can generally be categorized into 4 primary groups: 1) various player demographics and characteristics, 2) throwing too hard (high velocity), 3) throwing too much (pitch count/volume), and 4) throwing with poor mechanics. In this systematic review, the most significant nonmodifiable risk factors for MUCL injuries included: increased glenohumeral retrotorsion and elbow valgus opening. The most consistent modifiable risk factors included: total shoulder range of motion, pitch count, pitch selection, Y balance score, and lateral release position. Pitch velocity was inconsistent in literature, but most studies found this as a risk for injury. These risk factors may serve as appropriate targets for future evidence‐based injury mitigation strategies. Level of Evidence Level IV, systematic review of Level II‐IV studies.

Background: Prior studies have reported conflicting results of persistent iatrogenic atrial septal defect (iASD) and its impact following a transcatheter edge-to-edge repair (TEER) procedure. This study aims to evaluate the incidence of iASD and its clinical and hemodynamic impact after a TEER. Methods: Consecutive patients who underwent a TEER procedure from June 2014 to September 2020 at the Mayo Clinic were identified. The presence of iASD was retrospectively identified on post-procedure transthoracic echocardiography (TTE) to group patients into an iASD+ group and an iASD− group for comparison of prognosis and cardiac function. Results: A total of 316 patients were included; the mean age was 79.1 ± 9.1 years, and 67.7% were male. Persistent iASD was diagnosed in 108 (34.2%) patients. There was no difference concerning all-cause mortality, heart failure hospitalization, and stroke/TIA between groups at follow-up (median follow-up of 9 months). Post-procedure TTE demonstrated no differences regarding right ventricle (RV) and left ventricle (LV) dimensions and function, including TAPSE (15.2 ± 4.6 vs. 15.4 ± 5.5 mm, p = 0.875), and LV ejection fraction (51.1 ± 14.0% vs. 51.3 ± 13.9%, p = 0.933). However, patients with iASD had higher RV systolic pressure (48.7 ± 14.4 vs. 45.5 ± 14.5 mmHg, p = 0.042) compared with patients without iASD. Conclusion: Notwithstanding higher RV pressures, patients with persistent iASD had no hemodynamic compromise and an equal prognosis compared with those without a residual atrial defect after a TEER procedure. These findings support the mid-term safety of procedures in which an interatrial defect needs to be created and would argue against the need for interventional closure.

AbstractBackgroundFluorodeoxyglucose positron emission tomography (FDG‐PET) is an established modality for differential diagnosis of dementia. Deriving phenotypic signatures from FDG‐PET via machine learning is challenging due to the high dimensional nature of brain images relative to the generally small number of labeled examples available for training, the class imbalance among those labels, and the cooccurrence of multiple pathologies. In this study, we developed a multi‐class, multi‐label framework to address these challenges.MethodA database of clinically acquired PET/CT images from 3,000 unique patients was used to develop a latent space model using matrix decomposition. This model was then applied to images from a separate cohort of Mayo Clinic Alzheimer’s Disease Research Center participants (n=1,745) labeled as cognitively unimpaired (CU) (n=1,436) or with the following potentially co‐occurring phenotypes: Alzheimer’s disease (AD) (n=165), Lewy body dementia (DLB) (n=92), behavioral variant frontotemporal dementia (bvFTD) (n=43), semantic (svPPA) (n=10) and logopenic (lvPPA) (n=13) variant PPA, and posterior cortical atrophy (PCA) (n=17). A k‐nearest neighbors classifier that is robust to these imbalanced and overlapping labels was then trained on these examples. The resulting classifier was evaluated by area under receiver‐operator characteristic curve (ROC‐AUC) via leave one out cross validation, using clinical diagnosis as the gold standard.ResultROC curves and AUC scores for each phenotype are illustrated in Fig. 1a. Because the classifier is based on a k‐nearest neighbors connectivity matrix, it has a convenient graphical representation, where images are nodes and edges are drawn between an image and its set of nearest neighbors in latent space. A self‐organizing force directed graph constructed in this way is illustrated in Fig 1b, highlighting the strong separation of CU and degenerative images, as well as the segregation of each phenotype within the neurodegenerative region of the graph.ConclusionIn this study, we developed a machine learning framework for classification of neurodegenerative disease based on k‐nearest neighbor analysis in a low dimensional latent space projection of FDG‐PET images. By leveraging low‐dimensional representations and k‐nearest neighbors analysis, this framework is robust in multi‐class, multi‐label tasks with strong class imbalance and provides a highly interpretable graphical representation.

AbstractBackgroundIn vivo tau‐positron emission tomography (PET) is an attractive biomarker for Alzheimer’s disease (AD) diagnosis and treatment. However, tau‐PET is less widely available than other modalities. In this study, we tested cross‐modality synthesis of tau‐PET brain images from fluorodeoxyglucose F‐18 (FDG)‐PET using a deep convolutional neural network (CNN).MethodParticipants (n=1,192) who had brain FDG‐PET with 18F‐FDG and tau‐PET with Flortaucipir (F‐18‐AV‐1451) were included for training and testing. This cohort spanned normal aging (ages 26‐98), pre‐clinical, and clinical AD and related disorders including the FTD and DLB spectrum. External validation was done using ADNI (n=288). The PET scans were co‐registered to the corresponding MRI and subsequently warped to Mayo Clinic Adult Lifespan Template (MCALT) space. Tau‐PET images were SUVR‐normalized to the cerebellar crus, and FDG to the pons. A 3D dense‐U‐net model was utilized as an architecture. Cross‐validation experiments were conducted using 5‐fold validations (60% training set, 20% validation set, and 20% test set) with mean squared error as the loss function.ResultOur dense‐U‐net model successfully synthesized tau‐PET from metabolic images with good correlation and low prediction error for regional SUVRs (Figure 1A‐C). The model showed a robust prediction ability, performing accurately in an independent, external ADNI cohort (Figure 1D‐F). The model‐imputed tau‐PET significantly improved performance in classifying tau positivity (mean AUROC(±SD)=0.78±0.04 and 0.85±0.03 for FDG‐PET and synthesized tau‐PET, respectively) and diagnostic groups (cognitively unimpaired with abnormal amyloid‐PET vs. cognitively impaired with abnormal amyloid‐PET) compared to the original input FDG data (mean AUROC(±SD)=0.89±0.04, 0.85±0.05 0.91±0.04 for actual tau‐PET, FDG‐PET and synthesized tau‐PET, respectively) (Figure 2), suggesting enhanced clinical utility for metabolic images. The ADNI cohort also showed similar results (for tau positivity: AUROC=0.66 and 0.78 for FDG‐PET and synthesized tau‐PET, respectively; for CU A+ vs. CI A+: AUROC=0.86, 0.62, and 0.73 for actual tau‐PET, FDG‐PET, and synthesized tau‐PET; Figure 3).ConclusionWe showed that using a CNN model to predict tau‐PET from FDG‐PET is feasible. The synthesized tau‐PET can augment the value of FDG‐PET, facilitating the multi‐modal diagnosis of AD.

AbstractBackgroundDysexecutive Alzheimer's disease (dAD) is a relatively early‐onset variant of AD primarily degrading core executive functions in the absence of predominant behavioral symptoms. Clinical observations suggest substantial variability in clinical, cognitive and neuroimaging profiles within this syndrome. In this presentation, we will discuss our ongoing work disentangling this heterogeneity using unsupervised machine learning techniques.MethodWe collected clinical, neuropsychological and multimodal imaging (MRI, FDG‐PET, amyloid‐PET, tau‐PET) data on 52 dAD patients assessed in our behavioral neurology clinic at Mayo Clinic Rochester. We first performed a spectral clustering analysis based on FDG‐PET to delineate latent patterns of network degeneration (referred as “eigenbrains”) and assessed the relationships between these eigenbrains and clinical and cognitive symptomatology. We then performed a hierarchical clustering on these eigenbrains to derive data‐driven subtypes of dAD. We compared clinical and cognitive data between subtypes, and then compared the imaging profiles to 52 amyloid‐negative age‐ and sex‐matched controls.ResultSix eigenbrains explained approximatively 48% of the variance in FDG‐PET patterns and primarily reflected heteromodal to primary motor/sensory, left‐right hemispheric asymmetry, and anterior‐to‐posterior gradients of macro‐scale cortical organization (Fig 1). These eigenbrains differentially related to reported age at symptom onset, degree of clinical impairment, and performance on a wide range of cognitive domains (executive functions, episodic memory, and visuospatial). Hierarchical clustering revealed four dAD subtypes (diffuse, biparietal, left‐hemisphere, right‐hemisphere). These subtypes differed in reported age at symptom onset and cognitive profile, where the heteromodal‐diffuse subtype exhibited an overall worse clinical picture and the biparietal had a milder profile compared to other subtypes, which was not explained by disease duration. Additionally, spatial patterns of tau distribution and neurodegeneration overlapped with patterns of FDG‐PET hypometabolism in each subtype, whereas patterns of amyloid deposition were similar across subtypes (Figs 2 & 3).ConclusionAlmost half of the variance in macro‐scale patterns of hypometabolism in this dAD cohort was accounted for by six eigenbrains. These eigenbrains captured the inter‐individual variability in age at symptom onset and cognitive impairment. Four dAD subtypes derived from these eigenbrains revealed meaningful differences in clinical, cognitive, and imaging profiles. Recognizing this heterogeneity has significant clinical implications for diagnosis, prognosis, and symptom monitoring.

AbstractBackgroundFluorodeoxyglucose positron emission tomography (FDG‐PET) is an established modality for differential diagnosis of dementia. Deriving phenotypic signatures from FDG‐PET via machine learning is challenging due to the high dimensional nature of brain images relative to the generally small number of labeled examples available for training, the class imbalance among those labels, and the cooccurrence of multiple pathologies. In this study, we developed a multi‐class, multi‐label framework to address these challenges.MethodA database of clinically acquired PET/CT images from 3,000 unique patients was used to develop a latent space model using matrix decomposition. This model was then applied to images from a separate cohort of Mayo Clinic Alzheimer’s Disease Research Center participants (n=1,745) labeled as cognitively unimpaired (CU) (n=1,436) or with the following potentially co‐occurring phenotypes: Alzheimer’s disease (AD) (n=165), Lewy body dementia (DLB) (n=92), behavioral variant frontotemporal dementia (bvFTD) (n=43), semantic (svPPA) (n=10) and logopenic (lvPPA) (n=13) variant PPA, and posterior cortical atrophy (PCA) (n=17). A k‐nearest neighbors classifier that is robust to these imbalanced and overlapping labels was then trained on these examples. The resulting classifier was evaluated by area under receiver‐operator characteristic curve (ROC‐AUC) via leave one out cross validation, using clinical diagnosis as the gold standard.ResultROC curves and AUC scores for each phenotype are illustrated in Fig. 1a. Because the classifier is based on a k‐nearest neighbors connectivity matrix, it has a convenient graphical representation, where images are nodes and edges are drawn between an image and its set of nearest neighbors in latent space. A self‐organizing force directed graph constructed in this way is illustrated in Fig 1b, highlighting the strong separation of CU and degenerative images, as well as the segregation of each phenotype within the neurodegenerative region of the graph.ConclusionIn this study, we developed a machine learning framework for classification of neurodegenerative disease based on k‐nearest neighbor analysis in a low dimensional latent space projection of FDG‐PET images. By leveraging low‐dimensional representations and k‐nearest neighbors analysis, this framework is robust in multi‐class, multi‐label tasks with strong class imbalance and provides a highly interpretable graphical representation.

Abstract Characterizing intra- and inter-tumoral heterogeneity of glioblastoma has historically relied on discrete classifications of malignant cell populations leaving immune and other cell populations, known to exist admixed with the malignant tumor cells, relatively neglected. Manifold learning algorithms can manage deconvolving multiple cell populations and are often used to track cell state transitions in single cell transcriptomics. We applied a manifold learning approach to TCGA microarray data (Nf525) and bulk transcriptomics of 134 image localized biopsies across 30 patients with primary and 9 with recurrent glioblastoma to further elucidate how to organize biopsies across a spectrum of possible tissue states. The algorithm revealed a low-dimensional manifold graph for which each biopsy lives across 3 polarizing tissue states - one that is associated with diffusely invaded brain, one that is enriched in mesenchymal genes, and one that is enriched in classical proliferative tumor signatures. We deconvolved the bulk transcriptomics of the image-localized biopsies to reveal the relative abundance of 18 malignant, immune, and other cell subpopulations in each biopsy. Overlaying the cellular decomposition onto the manifold graph visualizing the tissue state distributions revealed that transitions between states correlate with changes in cellular cohabitation composition. The tumor cellular cohabitation ecologies have the lowest diversity, as inferred by ecological measures such as Shannon entropy and evenness, at the distal poles of the graph when compared to the transitional arms. Further, we found that the relationship between imaging appearance of contrast enhancement on T1-weighted MRI and the biopsy cellular composition varies with sex and primary vs recurrent biopsy status. The limited spectrum of possible tissue states revealed by the manifold learning is suggestive of a limited continuum along which tumor and non-tumoral cell populations can cohabitate. Such a limited low-dimensional biological space may constrain the dynamics of tumor biology in a predictable manner.